Examples

Transforming Features In The Iris Dataset

This is a version of part of the sklearn example Feature importances with forests of trees. It illustrates retaining the semantic meaning of features as they are transformed.

Feature Importance In The Iris Dataset

This is a version of the sklearn example Feature importances with forests of trees. It illustrates the consistent use of pandas data structures throughout the process.

Plotting Cross-Validated Predictions In The Boston Dataset

This is a version of the sklearn example Plotting Cross-Validated Predictions. It further illustrates the consistent use of pandas data structures and seaborn, as well as using ibex.xgboost.

Confidence Intervals In The Digits Dataset

This is a version of the sklearn example Pipelining: chaining a PCA and a logistic regression. It illustrates the use of pipelines.

Simple Row-Aggregating Features In The Movielens Dataset

This shows how to use pandas-munging estimators utilizing features which span multiple rows (instances). Pandas excels (no pun intended) in these kinds of operations.

Nonnegative Matrix Farcotization In The Movielens Dataset

This shows how to use pandas-munging estimators utilizing features which span multiple rows (instances), this time using nonnegative matrix factorization.

Tensorflow/Keras Classification In The Iris Dataset

This example shows how to use ibex.tensorflow.contrib.keras.wrappers.scikit_learn.KerasClassifier.